Europe focus

EPO developments on patentability of biotechnology inventions

By Rainer Friedrich and Sandra Pohlman, df-mp

This chapter explores developments with respect to European patents for biotechnological inventions, focusing on the patenting of antibodies, plants and stem cells.

Patenting antibodies

No specific directives govern the patentability of antibodies in Europe. The European Patent Office’s (EPO) requirements have instead evolved with the technology. Some of these requirements arise from decisions of the EPO Boards of Appeal, but EPO examiners are also guided by internal guidelines.

Definition by antigen

Antibody claims often contain features relating to antigen binding (eg, “an antibody that specifically binds to…”). If the antigen is new, an antibody against this antigen is usually considered to be novel and inventive, provided that the antigen is sufficiently defined in the application. The EPO considers it straightforward to obtain antibodies to a new protein and so does not require such antibodies to be exemplified.

Since disclosure of the antigen is considered as allowing the generation of an antibody to the antigen, claims relating to an antibody to a known antigen will usually lack inventive step, unless this antibody is shown to exhibit an advantageous property that distinguishes it from other antibodies against the same antigen, such as agonist rather than antagonist activity or unusually effective neutralisation of the antigen (T 645/02). This approach of the EPO may become problematic regarding prior art that describes an antibody prophetically (ie, without actually producing it), as it may be difficult in such circumstances to show an ‘unexpected property’ that supports inventive step.

Definition by epitope

In cases where the antigen to which the antibody binds is already known and some antibodies to that antigen have already been publicly disclosed, claims to antibodies that are directed to specific epitopes on that antigen might still be possible (assuming that the known antibodies are not directed to those epitopes). Epitopes may be defined by reference to a specific monoclonal antibody which binds to that epitope or by reference to the amino acid sequence of the epitope in the antigen. The scope of a patent claiming an antibody against a specific epitope of a given target will not cover antibodies against other epitopes of the same target; however, as this type of claim is more specific, it is often less susceptible to invalidity attacks.

The EPO may hold that it represents an undue burden for the skilled person to find additional antibodies which bind to the same epitope as the exemplified antibody. However, if at least one antibody which binds to the epitope is disclosed and the epitope has been characterised, there is a reasonable chance of overcoming this objection.

Definition by sequence

A claim may relate to the relatively exact structure of the antibody that inherently possesses the inventive function, by reference to the VH/VL domains or the entire antibody sequence. Traditionally, the EPO has accepted claims to antibodies defined by their six complementarity determining regions (CDRs), and many patents are granted on this basis (if the antibody has an unexpected property). Recent cases at the EPO indicate that reference in the claim to the six CDRs no longer suffices – the framework regions are considered to play a role in the binding affinity. Therefore, a claim specifying only the CDRs is considered to encompass embodiments where the affinity of the exemplified antibody is lost or likely to be lost.

Definition by function/activity

If the invention is based on the function of the antibody, it may be possible to claim the antibody using a functional feature without limitation to a particular sequence (T 2045/09), unless the property is already known from the prior art (T 735/00). In order to avoid undue burden objections, the application should provide an assay that can be used to screen for the activity.

Functional features may relate to:

  • competing for binding with another antibody;
  • cross-reacting with multiple antigens; or
  • selectively binding one antigen rather than another.

Another commonly claimed feature of functionally defined claims is the antibody’s reactivity or lack thereof. Other examples of antigen-related functional features include defined inhibitor/antagonist activities, biological effects in assays involving the target antigen and in vivo effects (T 2332/10).

Definition by a mix of structural and functional features

Claims relying on a structural antibody definition may provide relatively limited protection. A compromise could be to reduce the structural details in the claim and to add a functional definition that requires that the claimed antibody retain its function. A functional definition is arguably always required if the claim relates only to the six CDRs or even allows structural flexibility in the CDRs. Alternatively, data may show that fewer CDRs are required to achieve a technical effect or that the CDRs may be modified while maintaining a technical effect.

Definition by hybridoma

Antibodies can be defined with reference to a hybridoma deposit. However, thought should be given as to how variants of the antibody made by the hybridoma can be defined. It is possible that the examiner will require the claims to be restricted to the specific antibody made by the hybridoma. If an antibody is claimed by reference to a hybridoma, the deposit requirements pursuant to Rule 31 of the European Patent Convention (eg, depository name and accession number) should be met to avoid enablement issues. Preferably, the deposit should be made before the priority filing date and the deposit details should be given in the priority document.

Prior art antibodies may not yet have come to the applicant’s attention or a prior art antibody may exhibit a partial functional overlap. It may be useful to carry out structural and functional studies before filing a patent application for an antibody, so that relevant details can be included in the patent application. The application should allow flexibility for reformulating the problem by mentioning further properties such as antigen specificity, epitope binding, affinity/binding (Kon, Koff, Kd), neutralising titre (Ki), clearance rate, mechanism of action, antibody stability, catalytic activity or immunogenicity – even if no such data is available at the filing date. Evidence that routine methods would not have been sufficient or data showing that, for example, certain amino acid substitutions failed could also be helpful. The application should provide as much detail as possible for further limiting the specificity – for example, to exclude cross-reacting antibodies that bind to related proteins. This could be done by providing a basis for specifying that the antibody does not bind proteins with less than a given percentage homology to the protein that the antibody has been raised against.

Patenting plants

Plant varieties have historically been excluded from patentability to avoid overlap with plant variety rights, a separate form of intellectual property applicable to new varieties of plant. Such rights still have a valuable role, but some plant innovations are increasingly of a type to which patents are better suited. Article 53(b) of the European Patent Convention stipulates that European patents will not be granted in respect of “plant or animal varieties or essentially biological processes for the production of plants or animals”. The exclusion of ‘plant or animal varieties’ applies to specific, individually claimed varieties only. If a claim encompasses multiple plant varieties but does not cover any specific variety, the exclusion does not apply (G 1/98).

In its 2010 decisions Tomato I and Broccoli I (G 2/07 and G 1/08) the EPO distinguished between processes of genetic engineering – which do not fall within the exclusion – and conventional plant-breeding processes. For example, sexual crossing of whole haplotypes is an ‘essentially biological process’ and therefore excluded even if such crossing would not happen in nature – for example, due to the non-availability of starting material or a barrier to cross-pollination needing to be overcome by human intervention.

The reaction to these decisions was to claim plant inventions in product format. In Tomato II and Broccoli II (G 2/12 and G 2/13) the EPO held that subject matter claimed as a product or a product-by-process is not the same as one claimed for a process, which in the case of essentially biological processes for the production of plants is excluded from patentability – regardless of how the claimed product is generated or (as in the case of a product-by-process claim) defined. Even if the product (ie, the plant or plant material such as a fruit or plant parts) can be obtained only by essentially biological processes, with no other methods either disclosed in the patent application or otherwise known, the process exclusion in Article 53(b) of the European Patent Convention does not extend to product claims and product-by-process claims.

However, whether such product claims or product-by-process claims are allowable and lead to granting of a European patent depends on the fulfilment of the formal and substantive requirements of the European Patent Convention concerning these kinds of claim category (independent of the issue of the allowability of a patent claim directed to an essentially biological process for the production of plants). The allowability of a product-by-process claim is subject to the additional (restrictive) conditions established in the case law of the Boards of Appeal. Therefore, the mere fact that an applicant chooses a product claim or product-by-process claim instead of a method claim directed to an essentially biological process for the production of a plant is not a matter of skilful claim drafting or circumvention of legal hurdles, but a legitimate choice to obtain patent protection for the claimed subject matter, provided that the requirements for allowability are met.

Applicants should ensure that appropriate product language is included in the application, even if the focus in other territories may be on the process. Product claims may be allowable in certain circumstances, but product-by-process claims may be particularly useful where the end product cannot be easily defined. In the case of pending applications, an adequate basis for formulating product-by-process claims must be present in the original application.

Patenting stem cells

The patentability of biotechnological inventions in the European Union is governed by the EU Directive on the Patenting of Biotechnological Inventions (98/44/EC). One effect of the directive is to prohibit patents on the use of human embryos for commercial or industrial purposes. The EPO subsequently updated the European Patent Convention to include these rules.

Embryonic stem cells are derived from early-stage embryos and the original methods for obtaining such cells involved destroying the embryo. Recent decisions of the EPO and the European Court of Justice (ECJ) have considered what is meant by the term ‘embryo’ in this context and how this prohibition should be applied in the context of downstream products and processes.

In October 2011 the ECJ ruled that any invention involving the use of human embryonic stem cells (hESCs) is not patentable (C 34/10). The reasoning was that such cells are inherently derived from human embryos and as such constitute use of human embryos for industrial or commercial purposes. The ECJ held that it does not matter when the use of human embryos takes place in the context of the invention. Thus, it does not matter that a person could implement the invention without actually using an embryo – for instance, by purchasing an hESC from a stem cell bank.

In its 2014 decision in C 364/13, the ECJ held that a human embryo should cover only cells that have the inherent capacity to develop into a human being. The term ‘embryo’ therefore does not cover all cells that can start to develop into a human being – only those that can fully develop into a human being. Based on the evidence presented to the court, parthenogenetically activated human oocytes or other cell types that cannot fully develop into a human being may now be patentable in Europe.

In EPO Case T 2221/10 the claim did not require the destruction of a human embryo. The invention related to the maintenance of hESC in an undifferentiated state by the addition of certain human foreskin cells and to such a culture per se. The EPO considered that if such hESC were derived from human embryos that had been destroyed, it was irrelevant how early in the performance of the invention such destruction occurred. An applicant must prove that even if the hESC cell lines were publicly available, such cell lines have been prepared without destroying a human embryo. The evidence provided by the applicant suggested that embryos were indeed destroyed during the preparation of the stem cells.

The EPO refused an application directed to a method of producing differentiated cells from primate pluripotent stem cells, such as hESCs. The claimed method did not indicate the source of the cells, but the EPO concluded that the known and practised method of obtaining human embryonic stem cells at the filing date of the patent application required the destruction of human embryos. According to the evidence before the EPO, the first public disclosure of a method by which human embryonic stem cells could be obtained without destroying a human embryo was published in January 2008 (T 1441/13).

In practice, therefore, claims relating to human embryonic stem cells may be allowed by the EPO if they have a priority or filing date of January 2008 or later, as they can rely on published non-destructive methods of obtaining cells from human embryos – unless the application relates to particular cell lines that are known to have been produced by a destructive method. This date may shift if a date earlier than January 2008 can be established for the availability of stem cells produced by non-destructive methods.

If a claim encompasses both destructive and non-destructive methods, it may be necessary to disclaim any products, methods or uses that required the destruction of a human embryo (T 1808/13). Such a disclaimer can be used only in cases where non-destructive methods were available at the filing date (ie, cases filed in January 2008 or later).

Accordingly, the EPO excludes from patentability any invention necessarily involving the use and destruction of human embryos. Applicants should ensure that at least one example of either exclusively pluripotent cells, non-human embryonic stem cells or induced pluripotent stem cells is included in their patent application.


Patentanwälte Rechtsanwälte PartG mbB

Theatinerstrasse 16

80333 Munich


Tel +49 89 2102 960

Fax +49 89 2102 9633


Rainer Friedrich

Rainer Friedrich, a partner at df-mp, is authorised to represent clients before the European Patent Office, the German Patent and Trademark Office, the German Federal Patent Court and the Office for Harmonisation in the Internal Market. He has been active in the IP field since he joined df-mp in 2004.

Dr Friedrich’s scientific background in molecular biology, organic chemistry and biochemistry enhances the firm’s diversified life sciences practice group. His practice focuses on prosecution, opposition and litigation matters concerning German and European patents and supplementary protection certificates. Dr Friedrich has been involved in various high-level international patent disputes regarding protective rights in the above areas of technology. His practice also includes the provision of freedom-to-operate and validity opinions.

Sandra Pohlman

Sandra Pohlman is a co-founder and partner at df-mp. She heads the firm’s large and diversified biotechnology and pharmaceuticals practice group, together with co-founder H Ulrich Dörries. Ms Pohlman has specialised in intellectual property in the biotechnology and pharmaceutical fields since 1992.

Ms Pohlman advises domestic and foreign clients on all aspects of intellectual property in her field, with a particular focus on molecular biology, immunology, blood products, pharmaceutical agents and formulations, gene therapy, chromatography, diagnostic assays and medical applications.

Consistently (from 2013 to 2016), Ms Pohlman has been recommended for her expertise in patent prosecution and invalidity actions in the IAM Patent 1000. Managing Intellectual Property magazine listed her as an IP Star in 2014 and 2015.


Register for more free content

  • Read more IAM blogs and articles
  • Receive the editor's weekly review by email
Register now  
Issue 86